kepler

Energy estimator

An eBPF-based tool that collects system performance data and uses machine learning to estimate energy consumption in Kubernetes clusters.

Kepler (Kubernetes-based Efficient Power Level Exporter) uses eBPF to probe performance counters and other system stats, use ML models to estimate workload energy consumption based on these stats, and exports them as Prometheus metrics

GitHub

1k stars
23 watching
184 forks
Language: Go
last commit: 8 days ago
Linked from 3 awesome lists

cloud-nativeebpfenergy-consumptionenergy-efficiencyenergy-monitorkubernetesmachine-learningprometheusprometheus-exportersustainability

Backlinks from these awesome lists:

Related projects:

Repository Description Stars
hubblo-org/scaphandre A tool to measure and monitor the energy consumption of systems and applications. 1,639
responsibleproblemsolving/energy-usage Measures energy usage and carbon emissions of computational processes. 154
openeemeter/eemeter Provides tools for calculating energy savings based on standardized models of building energy usage. 220
cloudflare/ebpf_exporter An exporter for collecting custom eBPF metrics and OpenTelemetry traces from Linux kernel programs 2,200
treydock/prometheus-slurm-exporter Collects performance metrics from Slurm resource scheduling system and exports them as Prometheus metrics. 0
oemof/oemof-solph Generates models for energy system analysis and optimization. 303
kindlingproject/kindling An eBPF-based monitoring tool for Kubernetes applications. 1,122
green-coding-solutions/green-metrics-tool Tools to measure energy consumption of software throughout its life cycle 178
helmholtz-ai-energy/perun A Python package to measure and track the energy consumption of applications 52
openenergymonitor/emonpi An open-source energy monitoring platform built on Raspberry Pi and supported by a community-driven software stack 270
hebes-io/eensight A tool for measuring and verifying energy efficiency improvements in buildings 16
powerapi-ng/pyjoules A tool to measure energy consumption of code snippets by leveraging Intel RAPL and NVIDIA GPU technologies. 71
tricorder-observability/starship An Observability platform using eBPF and WASM to instrument and analyze system performance data 162
akoutmos/prom_ex A library for collecting and visualizing metrics from Elixir applications 616
grid-parity-exchange/egret A Python-based package for solving optimization problems in power systems 133